Neural Network For Pattern Classification Research Articles

A flexible enhanced fuzzy min-max neural network for pattern classification

In this paper, the existing enhanced fuzzy min–max (EFMM) neural network is improved with a flexible learning procedure for undertaking pattern classification tasks. Four new contributions are introduced. Firstly, a new training strategy is proposed for avoiding the generation of unnecessary overlapped regions between hyperboxes of different classes. The learning phase is simplified by eliminating the contraction procedure. Secondly, a new flexible expansion procedure is introduced. It eliminates the use of a user-defined parameter (expansion coefficient) to determine the hyperbox sizes. Thirdly, a new overlap test rule is applied during the test phase to identify the containment cases and activate the contraction procedure (if necessary). Fourthly, a new contraction procedure is formulated to overcome the containment cases and avoid the data distortion problem. Both the third and fourth contributions are important for preventing the catastrophic forgetting issue and supporting the stability-plasticity principle pertaining to online learning. The performance of the proposed model is evaluated with benchmark data sets. The results demonstrate its efficiency in handling pattern classification tasks, outperforming other related models in online learning environments.

A competitive learning scheme for deep neural network pattern classifier training

To reduce the computational complexity of training a deep neural network architecture using large data sets of 3D scenes, a competitive learning scheme was devised. The proposed algorithm pits a neural network learning algorithm, in this case the standard Adam optimiser, against an evolutionary algorithm that is used to select the most difficult training examples. The overall scheme is similar to a predator–prey system, where the predator (the neural network) strives to optimise its ability to capture (identify) the prey (the training patterns), and the evolutionary procedure selects the prey that so far evaded capture. As a consequence of the evolutionary process, the neural network is presented only a fraction of the training examples, and the computational complexity of the learning procedure is reduced. Experimental evidence showed that the proposed scheme allows reducing the deep neural network training time on different model sets, sometimes significantly, without affecting the recognition accuracy. The proposed predator–prey scheme is fairly independent of the ANN type and training algorithm employed, and has the potential to be beneficial to a wide range of deep learning applications, where practical implementations are often hindered by the time complexity of the training process.

Multi-component quantitative and feed-forward neural network for pattern classification of raw and wine-processed Corni Fructus

Multi-component quantitative and feed-forward neural network for pattern classification of raw and wine-processed Corni Fructus

Two phase learning technique in modular neural network for pattern classification of handwritten Hindi alphabets

Modular neural network overcomes the problem of monolithic structures of artificial neural networks. Generally modular neural network is an integration of smaller sub complete neural network models. Each model works independently on a sub portion of larger size pattern vectors. There are two ways of modularizing the neural network i.e. modularizing learning and modularizing structure. In this present work the modular neural network with modular learning for pattern classification of hand written Hindi alphabets is considered. In the presented approach 24 individual sub neural networks have been considered for first phase computing. In the second phase the collective outputs of first phase is presented as input to global neural network. Thus, the output of second phase presents the desired classification of the given large training set. Neural networks of first phase are trained locally for decomposed input patterns with gradient descent learning. Updated weights of the first phase are mapped to the global neural network. The global neural network is further trained for the collective output patterns of the first phase computing. Two phases of modular neural network i.e. decomposition and replication have been applied to perform the classification task. Simulation results are indicating that the complete neural network is approximated well when updated weights of first phase are combined with new weights of second phase and it generalized well when only updated weights of sub neural network of first phase are mapped to the connection strength of global neural network.

Passive Islanding Detection Based on Angular Velocity Harmonic Patterns with Perceptron Neural Network

Traditional islanding detection methods present improper action when the power mismatch between generation and load in the islanding system is small. The minimum power mismatch for correct islanding detection defines the limits os a Non-Detection Zone. This paper proposes an efficient method for islanding detection based on angular velocity harmonic patterns. To accomplish this, the measured frequency is decomposed by a Fourier Transform and applied to a Perceptron artificial neural network for pattern classification. The performance of the proposed method is evaluated by tests on a modified IEEE 34 node test system. Several islanding and non-islanding cases were simulated. The proposed method achived a performance of 88.16% in the classification of harmonic patterns with a small training set. Compared with the Under/Over frequency method, the proposed method represents a performance improvement of 17%.

A novel voting convergent difference neural network for diagnosing breast cancer

Breast cancer is one of the most frequently occurred cancers for females, and thus diagnosing breast cancer is very important. Neural dynamic algorithm (NDA) has been successfully applied in many fields with the characteristics of parallel computing and exponential convergence. However, there is no research on applying NDA-based neural network for pattern classification. In this paper, a novel voting convergent difference neural network (V-CDNN) is proposed. To do so, samples are firstly handled by feature selection, feature weighting and sample normalization. Secondly, the preprocessed samples are used to simultaneously and independently train several convergent difference neural networks in different types of mapping functions. Thirdly, in the testing process, voting strategy for these networks is applied to make diagnosis results more accurate and convincing. Being different from most existing neural networks, the proposed V-CDNN adopts neural dynamic learning algorithm, which greatly improves computation efficiency and increases accuracy rate of diagnosis. Experimental results verify that the proposed V-CDNN can achieve 100% average diagnosis accuracy, which is the highest among existing state-of-the-art methods on the open data set.

Variational Learning for Quantum Artificial Neural Networks

In the last few years, quantum computing and machine learning fostered rapid developments in their respective areas of application, introducing new perspectives on how information processing systems can be realized and programmed. The rapidly growing field of Quantum Machine Learning aims at bringing together these two ongoing revolutions. Here we first review a series of recent works describing the implementation of artificial neurons and feed-forward neural networks on quantum processors. We then present an original realization of efficient individual quantum nodes based on variational unsampling protocols. We investigate different learning strategies involving global and local layer-wise cost functions, and we assess their performances also in the presence of statistical measurement noise. While keeping full compatibility with the overall memory-efficient feed-forward architecture, our constructions effectively reduce the quantum circuit depth required to determine the activation probability of single neurons upon input of the relevant data-encoding quantum states. This suggests a viable approach towards the use of quantum neural networks for pattern classification on near-term quantum hardware.

(Invited) Computing with Memristive Devices and Arrays

Memristive devices have become a promising candidate for unconventional computing due to their attractive properties(1). The computing can be implemented on a Resistive Neural Network (ResNN) with memristor synapses and neurons or a Capacitive Neural Network (CapNN) with memcapacitor synapses and neurons(2).For ResNNs as computing accelerators, we have built a dot-product engine based on a 128 x 64 1T1R crossbar array using traditional non-volatile memristors with 64 stable analog resistance levels(3). With such computation accelerators, we have demonstrated efficient inference and learning with traditional Machine Learning algorithms(4-7), which is expected to significantly improve the speed and energy efficiency of neural networks.For ResNNs beyond accelerator applications, we developed diffusive memristors(8)with diffusion dynamics that is critical for neuromorphic functions. Based on the diffusive memristors, we have further developed artificial synapses(8)and neurons(9)to more faithfully emulate their bio-counterparts. We then integrated these artificial synapses and neurons into a small neural network, with which pattern classification and unsupervised learning have been demonstrated(9).For CapNNs, we have developed pseudo-memcapacitive devices based on the diffusive memristors. Capacitive synapses and neurons enabled by these memcapacitive devices have been developed and used to form a fully integrated CapNN(10), which can implement spiking signal classification and Hebbian-like learning. Z. Wanget al., Resistive switching materials for information processing. Nature Reviews Materials 5, 173–195 (2020). Q. Xia, J. J. Yang, Memristive crossbar arrays for brain-inspired computing. Nature materials 18, 309-323 (2019). C. Liet al., Analogue signal and image processing with large memristor crossbars. Nature Electronics 1, 52-59 (2018). Z. Wanget al., Reinforcement learning with analogue memristor arrays. Nature Electronics 2, 115-124 (2019). Z. Wanget al., In situ training of feed-forward and recurrent convolutional memristor networks. Nature Machine Intelligence 1, 434-442 (2019). C. Liet al., Long short-term memory networks in memristor crossbar arrays. Nature Machine Intelligence 1, 49-57 (2019). P. Linet al., Three-dimensional memristor circuits as complex neural networks. Nature Electronics 3, 225–232 (2020). Z. Wanget al., Memristors with diffusive dynamics as synaptic emulators for neuromorphic computing. Nature Materials 16, 101-108 (2017). Z. Wanget al., Fully memristive neural networks for pattern classification with unsupervised learning. Nature Electronics 1, 137-145 (2018). Z. Wanget al., Capacitive neural network with neuro-transistors. Nature Communications 9, 3208 (2018).

Advanced Deep Learning Spectroscopy of Scalogram Infused CNN Classifiers for Robust Identification of Post‐Hypoxic Epileptiform EEG Spikes

There is a lack of reliable prognostic biomarkers for hypoxic‐ischemic (HI) brain injury in preterm infants. Herein, spectrally detailed wavelet scalograms (WSs), derived from the 1024 Hz sampled electroencephalograms (EEG) of preterm fetal sheep after HI (n = 7), are infused into a high‐performance deep convolutional neural network (CNN) pattern classifier to identify high‐frequency spike transient biomarkers. The deep WS‐CNN pattern classifier identifies EEG spikes with remarkable accuracy of 99.81 = 0.15% (area under curve, AUC = 1.000), cross‐validated across 5010 EEG waveforms, during the first 6 h post‐HI (42 h total), an important clinical period for diagnosis of HI brain injury. Further, a feature‐fusion strategy is introduced to extract the spectrally dominant features of the raw EEG epochs to form robust 3D input matrix sets to be infused into the deep 2D‐CNNs for pattern classification. The results show that the proposed WS‐CNN approach is less sensitive to the potential morphological variations of spikes across all subjects compared to other deep CNNs and spectral‐fuzzy classifiers, allowing the user to flexibly choose an approach depending on their computational requirements. Collectively, the data provide a reliable framework that could help support well‐timed diagnosis of at‐risk neonates in clinical practice.

A Bilevel Learning Model and Algorithm for Self-Organizing Feed-Forward Neural Networks for Pattern Classification.

Conventional artificial neural network (ANN) learning algorithms for classification tasks, either derivative-based optimization algorithms or derivative-free optimization algorithms work by training ANN first (or training and validating ANN) and then testing ANN, which are a two-stage and one-pass learning mechanism. Thus, this learning mechanism may not guarantee the generalization ability of a trained ANN. In this article, a novel bilevel learning model is constructed for self-organizing feed-forward neural network (FFNN), in which the training and testing processes are integrated into a unified framework. In this bilevel model, the upper level optimization problem is built for testing error on testing data set and network architecture based on network complexity, whereas the lower level optimization problem is constructed for network weights based on training error on training data set. For the bilevel framework, an interactive learning algorithm is proposed to optimize the architecture and weights of an FFNN with consideration of both training error and testing error. In this interactive learning algorithm, a hybrid binary particle swarm optimization (BPSO) taken as an upper level optimizer is used to self-organize network architecture, whereas the Levenberg-Marquardt (LM) algorithm as a lower level optimizer is utilized to optimize the connection weights of an FFNN. The bilevel learning model and algorithm have been tested on 20 benchmark classification problems. Experimental results demonstrate that the bilevel learning algorithm can significantly produce more compact FFNNs with more excellent generalization ability when compared with conventional learning algorithms.

Artificial Neural Network–Based Multisensor Monitoring System for Collision Damage Assessment of Lithium‐Ion Battery Cells

The sales of electric vehicles (EVs) have seen a significant upward trend in recent years. The occurrence of severe collision accidents has raised awareness concerning safety issues of lithium‐ion batteries (LIBs) which are the core components of EV energy storage systems. When collision accidents of vehicles occur, in most cases, the battery cells will immediately burn and even explode, whereas in some cases, despite not directly showing visual damages, the battery cells may lead to delayed catastrophic failures. Herein, the assessment of the battery cells collision damage based on sensor signal data is focused on by identifying potentially unsafe cells. A campaign of collision experimental tests and a series of electrical performance tests are conducted on single‐cell specimens. A cell damage characterization procedure is proposed and implemented on the battery cells after the collision tests. The impact force and z‐axis acceleration signal features and respective damage classes are input to an artificial neural network (ANN) pattern classifier to train a model to assess the battery cell collision damage.

A Monolayer Leaky Integrate‐and‐Fire Neuron for 2D Memristive Neuromorphic Networks

Abstract2D material based memristors have exhibited superior performance as artificial synapses for neuromorphic computing. However, 2D artificial neurons as have note been exploited as an indispensable computational element owing to the rich dynamics of neurons, which impede the construction of a 2D neuromorphic network. A methodology is developed by introducing ionic migration dynamics and electrochemical reaction into monolayer MoS2 single crystal and a 2D artificial neuron is realized. The sophisticated electrophysiology process of leaky integrate‐and‐fire (LIF) is emulated by the injection and extraction of Ag+ ions under an e‐field in a monolayer MoS2 device with fine‐tuned channel length. Moreover, the fire frequency and relaxation time of the artificial neurons can be readily modulated by adjusting the input voltage pulses. By directly capturing conductive filament under a scanning electron microscope, the underlying mechanism of the unique resistive switching of the 2D artificial neuron is attributed to the rapid diffusion and migration of Ag in MoS2 lattice under the e‐field. The formation and rupture of the conductive Ag filament enable volatile resistive switching and LIF behaviors. Furthermore, MoS2‐based neurons are integrated with a nonvolatile synapse array to build a full memristive artificial neural network and implement pattern classification, paving the way for the construction of 2D neuromorphic networks.

Fuzzy Neural Network for Pattern Classification

This research work proposes a fuzzy neural network (FNN) for pattern classification. The proposed network is the modified version of the Radial basis function neural network (RBFNN). FNN uses supervised fuzzy clustering and pruning algorithm to determine the precise number of clusters with proper centroid and width to form the processing nodes in the hidden layer. These clusters represent fuzzy set hyperspheres (FSHs), which are defined by the fuzzy membership function. The training between the hidden layer to output layer which is done by using the LMS algorithm in RBFNN is avoided, and the output is determined by using the fuzzy union operation. The fuzzy membership function shields the clustered patterns resulting in 100% accuracy for the data set used during training. Unlike other clustering algorithms used to construct the hidden layer of RBFNN, the proposed clustering algorithm is independent of tuning parameters and is fast in training and retrieval. Thus FNN reduces the computation time, guarantees 100% accuracy for any training set, and provides superior and comparable recognition accuracy for the datasets with the precise number of FSHs in the hidden layer. Hence the proposed FNN can be used for pattern classification.

(Invited) Computing with Memristive Devices and Arrays

Memristive devices1have become a promising candidate for unconventional computing2due to their attractive properties3. The computing can be implemented on a Resistive Neural Network (ResNN) with memristor synapses and neurons or a Capacitive Neural Network (CapNN) with memcapacitor synapses and neurons. For ResNNs as computing accelerators, we have built a dot-product engine based on a 128 x 64 1T1R crossbar array using traditional non-volatile memristors with 64 stable analog resistance levels4. Accurate image compression and filtering have been demonstrated with such analog computing accelerator4. In addition, we have demonstrated efficient and self-adaptive in-situlearning in a two-layer neural networks using such memristive arrays5, which is expected to significantly improve the speed and energy efficiency of machine learning. For ResNNs beyond accelerator applications, we developed diffusive memristors6with diffusion dynamics that is critical for neuromorphic functions. Based on the diffusive memristors, we have further developed artificial synapses7and neurons8to more faithfully emulate their bio-counterparts. We then integrated these artificial synapses and neurons into a small neural network, with which pattern classification and unsupervised learning have been demonstrated8. For CapNNs, we have developed pseudo-memcapacitive devices based on the diffusive memristors. Capacitive synapses and neurons enabled by these memcapacitive devices have been developed and used to form a fully integrated CapNN9, which has shown spiking signal classification and Hebbian-like learning functions. Moreover, the memristors can be used for cybersecurity10,11, LSTM12and robotics applications13. 1 Xia, Q., Yang, J. J., Memristive crossbar arrays for bio-inspired computing, Nature Materials 18 (2019). 2 Yang, J. J., Strukov, D. B. & Stewart, D. R. Memristive devices for computing. Nature Nanotechnology 8, 13 (2013). 3 Pi, S.et al.Memristor crossbar arrays with 6-nm half-pitch and 2-nm critical dimension. Nature Nanotechnology 14, 35 (2018). 4 Li, C.et al.Analogue signal and image processing with large memristor crossbars. Nature Electronics 1, 52 (2017). 5 Li, C.et al.Efficient and self-adaptive in-situ learning in multilayer memristive neural networks. Nature communications 9, 2385 (2018). 6 Midya, R.et al.Anatomy of Ag/Hafnia-Based Selectors with 1E10 Nonlinearity. Advanced Materials 29, 1604457 (2017). 7 Wang, Z.et al.Memristors with diffusive dynamics as synaptic emulators for neuromorphic computing. Nature Materials 16, 101 (2017). 8 Wang, Z.et al.Fully memristive neural networks for pattern classification with unsupervised learning. Nature Electronics 1, 137 (2018). 9 Wang, Z.et al.Capacitive neural network with neuro-transistors. Nature Communications 9, 3208 (2018). 10 Jiang, H.et al.A Novel True Random Number Generator Based on a Stochastic Diffusive Memristor.Nature communications 8, 882 (2017). 11 Jiang, H.et al.Provable Key Destruction with Large Memristor Crossbars. Nature Electronics 1, 548 (2018). 12 Li, C.et al.Long short-term memory networks in memristor crossbars. Nature Machine Intelligence 1, 49 (2019). 13 Yoon, J. H.et al.An artificial nociceptor based on a diffusive memristor. Nature Communications 9, 417 (2018).

A Critical Review on Selected Fuzzy Min-Max Neural Networks and Their Significance and Challenges in Pattern Classification

At present, pattern classification is one of the most important aspects of establishing machine intelligence systems for tackling decision-making processes. The fuzzy min–max (FMM) neural network combines the operations of an artificial neural network and fuzzy set theory into a common framework. FMM is considered one of the most useful neural networks for pattern classification. This paper aims to 1) analyze the FMM neural network in terms of its impact in addressing pattern classification problems; 2) examine models that are proposed based on the original FMM model (i.e., existing FMM-based variants); 3) identify the challenges associated with FMM and its variants, and; 4) discuss future trends and make recommendations for improvement. The review is conducted based on a methodical protocol. Through a rigorous searching and filtering process, the relevant studies are extracted and comprehensively analyzed to adequately address the defined research questions. The findings indicate that FMM plays a critical role in providing solutions to pattern classification issues. The FMM model and a number of FMM-based variants are identified and systematically analyzed with respect to their aims, improvements introduced and results achieved. In addition, FMM and its variants are critically analyzed with respect to their benefits and limitations. This paper shows that the existing FMM-based variants still encounter issues in terms of the learning process (expansion, overlap test, and contraction), which influence the classification performance. Based on the review findings, research opportunities are suggested to propose a new model to enhance the number of existing FMM models, particularly in terms of their learning process by minimizing hyperbox overlap pertaining to different classes as well as avoiding membership ambiguity of the overlapped region. In short, this review provides a comprehensive and critical reference for researchers and practitioners to leverage FMM and its variants for undertaking pattern classification tasks.

Modified Weights-and-Structure-Determination Neural Network for Pattern Classification of Flatfoot

Modified Weights-and-Structure-Determination Neural Network for Pattern Classification of Flatfoot

(Invited) Unconventional Computing with Memristive Devices and Arrays

Memristive devices1 have become a promising candidate for unconventional computing2. In this talk, I will present some of our recent work on unconventional computing experimentally implemented by using memristive devices or crossbar arrays. Using traditional non-volatile memristors with 64 stable analog resistance levels, we have built a dot-product engine based on a 128 x 64 1T1R crossbar array3. Accurate image compression and filtering have been demonstrated with such analog computing accelerator3. In addition, we have demonstrated efficient and self-adaptive in-situ learning in a two-layer neural networks using such memristive arrays4, which is expected to significantly improve the speed and energy efficiency of deep neural networks. Using our newly developed diffusive memristors5 with diffusion dynamics that is critical for neuromorphic functions, we have developed artificial synapses6 and neurons7 to more faithfully emulate their bio-counterparts and more efficiently perform spiking neural network functions. We have further integrated these artificial synapses and neurons into a small neural network, with which pattern classification and unsupervised learning have been demonstrated7. Moreover, the diffusive memristors can be used as true random number generators8 for cybersecurity applications and artificial nociceptors for robotics applications9. 1. Yang, J. J. et al. Memristive switching mechanism for metal/oxide/metal nanodevices. Nature Nanotechnology 3, 429 (2008). 2. Yang, J. J., Strukov, D. B. & Stewart, D. R. Memristive devices for computing. Nature Nanotechnology 8, 13 (2013). 3. Li, C. et al. Analogue signal and image processing with large memristor crossbars. Nature Electronics 1, 52 (2018). 4. Li, C. et al. Efficient and self-adaptive in-situ learning in multilayer memristive neural networks. Nature communications, under revision (2018). 5. Midya, R. et al. Anatomy of Ag/Hafnia-Based Selectors with 1E10 Nonlinearity. Advanced Materials 29, 1604457 (2017). 6. Wang, Z. et al. Memristors with diffusive dynamics as synaptic emulators for neuromorphic computing. Nature Materials 16, 101-108 (2017). 7. Wang, Z. et al. Fully memristive neural networks for pattern classification with unsupervised learning. Nature Electronics 1, 137 (2018). 8. Jiang, H. et al. A Novel True Random Number Generator Based on a Stochastic Diffusive Memristor. Nature communications 8, 882 (2017). 9. Yoon, J. H. et al. An artificial nociceptor based on a diffusive memristor. Nature Communications 9, 417 (2018).

Fusion of progressive granular neural networks for pattern classification

Evolving granular neural network (eGNN) is a system that classifies the data by adapting its architecture according to the input information. eGNN posses the ability to learn from the data with a single pass. However, eGNN has the disadvantage of linear updation of weights during training stage that may not handle the real-time uncertain and imprecise data appropriately. To address this issue, the present study aims to use multiple numbers of progressive granular neural networks (PGNNs) in the framework of fusion of classification systems. PGNN is a modified version of eGNN, where the learning process is nonlinear and takes the class-sensitive (CS) granulated data unlike eGNN. Collective opinion from the group of PGNNs increases the classifier performance in comparison with individual PGNNs. The proposed model thus takes the advantages of CS granulation, structural adaptability, nonlinear updation of weights and collective opinions from the group of PGNNs. Performance of model is tested on various datasets, and its superiority to similar other methods has been justified. Various performance indices, such as overall accuracy, dispersion score, kappa coefficient, producer’s accuracy, and user’s accuracy, have been used for performance analysis.

Fully memristive neural networks for pattern classification with unsupervised learning

Neuromorphic computers comprised of artificial neurons and synapses could provide a more efficient approach to implementing neural network algorithms than traditional hardware. Recently, artificial neurons based on memristors have been developed, but with limited bio-realistic dynamics and no direct interaction with the artificial synapses in an integrated network. Here we show that a diffusive memristor based on silver nanoparticles in a dielectric film can be used to create an artificial neuron with stochastic leaky integrate-and-fire dynamics and tunable integration time, which is determined by silver migration alone or its interaction with circuit capacitance. We integrate these neurons with nonvolatile memristive synapses to build fully memristive artificial neural networks. With these integrated networks, we experimentally demonstrate unsupervised synaptic weight updating and pattern classification.

Improved Learning in Constructive Neural Networks for Pattern Classification using Genetic Algorithm

Objective: To remove rogue entries in the data set chose for training thus making the Neural Network (NN) learning easier. Methods/Analysis: In this research work Genetic Algorithm (GA) is being used on the data sets Wine, Pima Indians Diabetes, Iris, Vehicle and Image Segmentation from the UC Irvine machine learning repository. The refined data sets are then fed to a Constructive Neural Network (CoNN) for pattern classification task. CoNN dispenses an optimal strategy in designing the structures of the multilayer perceptron networks, coupled with supervised algorithms for further learning. They eliminate the requisite of ad-hoc, apriori design of network architecture which is often inappropriate. The M-Tiling algorithm is used for constructing the network and a simple perceptron learning rule for training individual TLUs (Threshold Logic Unit). The CoNNs potential capability is in constructing networks with a network size adequate to the complexity level of the task and in attaining a satisfactory level of efficiency. Findings: Acquiring data from real world problems may result in the frequent occurrence of rogue values. Rogue values result in poor data quality and mislead the learning curve yielding imprecise NN models. A template string prototype is generated with the continual application of GA on the input data. This template string prototype removes the rogue values from the actual training set which results in the reduction of the size of input data set, in turn, makes learning easier. CoNN coupled with GA classifies patterns in Wine, Pima Indians Diabetes, Iris, Vehicle and Image Segmentation datasets available in UC Irvine machine learning repository with faster convergence, more generalization accuracy, and less space. The proposed work is unique since GA is exerted to the training data set before it is fed to the CoNN for learning. The proposed work also demonstrates the improvement of CoNN in terms of convergence, generalization accuracy, and space optimization. In this way, these experimental results add value to the currently existing work. Improvements: The proposed model can be further revamped by using other CoNN learning strategies.